This is a revised application in response to PAR-11-203: Predictive Multiscale Models for Biomedical, Biological, Behavioral, Environmental and Clinical Research (Interagency U01). Analyzing the Cardiac Power Grid: Adaptive responses to fluctuating local demand brings together seasoned investigators at U Washington, UC San Diego, University of Michigan, and Medical College of Wisconsin. This project will provide a central, core structure for integrative thermodynamically- and anatomically-constrained versions of the Cardiac Physiome, a large multiscale, multi-modular computer representation and resource for collating, coalescing and conciliating models as hypotheses and data from all 4 groups as tests of their validity. The research products will be publicly available, comprehensive data and models systematizing these data from subcellular, cell, tissue and organ levels: these will guide further experimental studies to test hypotheses and to identify therapeutic targets. The models will be composed of reproducible modules, each modifiable to accommodate new knowledge, and each in several forms ranging from simple descriptive to deeply mechanistic. The approach is middle out, neither top-down, nor bottom-up, but cell-centric, covering interactions among several cell types, with information flow up to the organ level and down to the protein level, so providing the sites for future linkages to regulate transcription. The data, he tools developed, the modules and the composite models, and the reproducible exchange packages (REP) of data and models are all open source. The three aims span Cell, Tissue, and Organ levels, and are strongly intertwined: (Aim 1) Power Generation and its Distribution on the Grid focuses on cellular and mitochondrial energetics, substrates for energy supply and purine nucleoside transport and exchange. It is given the largest write-up and is used as the power source by Aims 2 and 3. (Aim 2) Heterogeneities in Power Dissipation: Local Workloads links anatomic (high-resolution ventricular fiber and sheet arrangements), physiologic (excitatory spread and force generation) and metabolic (ATP and substrate use) to relate to regional myocardial blood flows. (Aim 3) Cardiomyocyte Control of Flows of Substrates for Power Generation involves 4 cell types (cardiomyocyte, smooth muscle, endothelium and RBC) and the relationships amongst them controlling blood flow. (Aim 4) Simulation Resource Core provides tools for modeling analysis of data, develops new technology for the semi-automated assembly, and disassembly and modification of models using reusable modules. Annotation of model code against standard ontologies identifies model elements precisely, and facilitates reuse by researchers worldwide.